11. Multiobjective De Novo Design of Synthetically Accessible Compounds

  1. Gisbert Schneider
  1. Valerie J. Gillet1,
  2. Michael J. Bodkin2 and
  3. Dimitar Hristozov2

Published Online: 11 OCT 2013

DOI: 10.1002/9783527677016.ch11

De novo Molecular Design

De novo Molecular Design

How to Cite

Gillet, V. J., Bodkin, M. J. and Hristozov, D. (2013) Multiobjective De Novo Design of Synthetically Accessible Compounds, in De novo Molecular Design (ed G. Schneider), Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, Germany. doi: 10.1002/9783527677016.ch11

Editor Information

  1. ETH Zürich, Institute of Pharmaceutical Sciences, Wolfgang-Pauli-Strasse 10, 8093 Zürich, Switzerland

Author Information

  1. 1

    Information School, University of Sheffield, Regent Court, 211 Portobello Street, Sheffield, S1 4DP, UK

  2. 2

    Medicinal Chemistry, Eli Lilly UK, Erl Wood Manor, Windlesham, Surrey GU20 6PH, UK

Publication History

  1. Published Online: 11 OCT 2013
  2. Published Print: 13 NOV 2013

ISBN Information

Print ISBN: 9783527334612

Online ISBN: 9783527677016



  • de novo design;
  • structure generation;
  • reaction vectors;
  • synthetic accessibility;
  • multiobjective optimization;
  • KNIME;
  • evolutionary algorithms;
  • generated databases


The first programs for the de novo design of molecules as potential drug candidates appeared more than 20 years ago; however, there was limited uptake at that time. A major factor preventing their use was the synthetic intractability of many of the compounds that were designed. Recently, there has been renewed interest in de novo design with attention focused on designing compounds that are synthetically accessible. Furthermore, the high rates of attrition seen in late stages of the drug discovery pipeline have resulted in a greater emphasis being placed on designing compounds that are optimized on their physicochemical properties as well as predicted potency. This chapter introduces the challenges in de novo design and then describes recent progress toward the design of molecules that are synthetically accessible. This is followed by an overview of multiobjective optimization techniques that enable multiple properties to be satisfied simultaneously. Finally, some recent approaches to de novo design that consider these two factors are described.